remove the block fill rate limit of 70% when farming a block

[arvidn]

Purpose:

to increase transaction rate

Current Behavior:

Farmers only fill blocks to 70% full

New Behavior:

Farmers fill blocks all the way. But blocks are compressed and have potential to fit more transactions.

Testing Notes:

[coveralls-official]

Pull Request Test Coverage Report for Build 12281997874

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 17 of 17 (100.0%) changed or added relevant lines in 2 files are covered.
• 547 unchanged lines in 23 files lost coverage.
• Overall coverage increased (+0.01%) to 91.482%

---

Files with Coverage Reduction	New Missed Lines	%
chia/timelord/timelord.py	1	80.0%
chia/full_node/pending_tx_cache.py	1	96.55%
chia/_tests/wallet/test_transaction_store.py	1	99.58%
chia/wallet/util/wallet_sync_utils.py	1	86.07%
chia/daemon/client.py	1	74.58%
chia/server/rate_limit_numbers.py	1	97.78%
chia/_tests/core/full_node/stores/test_full_node_store.py	1	99.5%
chia/_tests/core/full_node/stores/test_block_store.py	2	99.53%
chia/_tests/pools/test_pool_rpc.py	3	99.07%
chia/wallet/wallet_node.py	3	88.1%

Totals
Change from base Build 12209661102:	0.01%
Covered Lines:	104661
Relevant Lines:	114216

---

💛 - Coveralls

[emlowe]

do you want to comment what this magic number is

[arvidn]

yeah, it's not great. it's the number of transactions we need to fill the mempool sufficiently to trigger the expected number of rejections. going from fill rate 70% -> 100%, I simply took 169 / 0.7 to increase this number proportionally. I'll try to improve this a bit.

[AmineKhaldi]

Sent #19026 as an attempt to clarify this.

[emlowe]

I think I might prefer the simpler construct with a comment like Amine has above. The latest version here Arvid has more magic numbers :-)

[arvidn]

I think magic numbers are "more magic" the harder they are to explain. And breaking this down to how the magic number is computed makes the new ones less magical. (and I think in the most recent update there no magic numbers). As long as you accept that the CLVM cost of returning a quoted list is 44 and that it takes 88 bytes to serialize another agg-sig condition.

There are a lot of mempool tests with magic number that need tweaking when tweaking constants. Now we're moving this test to have its expectations properly rooted in the test constants it's using. I think this really helps humans understand (and gain confidence in) the test. I will follow up (in a separate PR) to make sure this test